A method and system for applying organic fertilizer in pasture planting based on soil fertility prediction

Abstract

The application discloses a kind of pasture planting organic fertilizer application method and system based on fertility prediction, specifically related to the technical field of fertility prediction using digital data processing, for solving the problem of calculation deviation of organic fertilizer application amount caused by the fact that the nutrient release of previous crop root stubble decomposition and soil background mineralization nutrient cannot be distinguished in multiple-crop pasture planting;The application obtains the daily variation sequence of soil cellulase activity and identifies the active change point, synchronously obtains the soil available nutrient time series monitoring value of planting area and root stubble residue-free control area from the active change point, obtains the difference curve through slope ratio judgment and difference operation, obtains the time point corresponding to the corrected initial rising slope and curvature maximum value based on the difference curve, separates endogenous decomposition release amount and soil background nutrient holding amount through source analysis relationship, determines the minimum application amount combined with the mineralization release characteristics of bedding biological organic fertilizer and applies, realizes double deduction of endogenous nutrient and background nutrient and precise application.

Description

Technical Field

[0001] This invention relates to the technical field of fertility prediction using digital data processing, and more specifically, to a method and system for applying organic fertilizer to forage crops based on fertility prediction. Background Technology

[0002] High-quality forage grasses are typically harvested three to five times a year. During cultivation, fermented bedding substrate and bio-organic fertilizer are commonly used as both base fertilizer and top dressing. This type of organic fertilizer originates from the cycle of straw bedding, fermented bedding farming, and bedding composting. Its nutrients are primarily in organic form, and after being applied to the soil, they gradually release available nitrogen, phosphorus, and potassium for forage grass absorption through continuous microbial mineralization. After each forage grass harvest, growers determine the amount of organic fertilizer to apply for the next harvest based on soil test results or experience, aiming to maintain a balance between soil fertility and forage grass growth needs. Accurate assessment of soil fertility trends is crucial for determining the appropriate application rate. Existing fertility prediction methods typically treat each fertilization as an independent event, using the measured soil nutrient values ​​after the previous harvest as the initial state for the next harvest prediction, and combining these with the mineralization rate parameters of the organic fertilizer and the target forage grass yield to estimate the fertilizer requirement.

[0003] However, under multiple-crop harvesting systems, there is a systematic bias between the fertility assessment results obtained by the above prediction methods and the actual fertility supply capacity. A large amount of root stubble and fallen leaves remaining in the soil after harvest decompose rapidly in a short period, releasing considerable mineral nutrients. These nutrients originate from the return of the previous crop's own biomass, not from the newly applied organic fertilizer. Existing technologies, when dealing with fertility prediction during crop rotation periods, fail to distinguish between the decomposition and release process of previous crop residues and the mineralization process of the current season's organic fertilizer. This leads to confusion in identifying the sources of available soil nutrients, causing the prediction results to overestimate the current season's contribution of organic fertilizer. Consequently, this results in a cumulative bias in the calculation of recommended application rates for subsequent crops, affecting fertilizer utilization efficiency and yield stability in multiple-crop forage planting systems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for applying organic fertilizer to forage grass planting based on fertility prediction to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for applying organic fertilizer to forage crops based on fertility prediction includes: S1: Obtain the diurnal variation sequence of soil cellulase activity after harvesting in the target plot; S2: Identify whether there are activity jump points in the daily activity variation sequence that exceed the range of daily average background fluctuations; S3: If it exists, then starting from the time corresponding to the active jump point, obtain the soil effective nutrient time-series monitoring value of the planting area as the first time-series monitoring value, and obtain the soil effective nutrient time-series monitoring value of the control area without root stubble residue as the second time-series monitoring value. S4: Calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, subtract the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value to obtain the difference curve. Based on the difference curve, obtain the corrected initial rising slope and the time point corresponding to the maximum curvature. S5: Input the corrected initial upward slope and the time point corresponding to the curvature maximum value into the source analysis relation to obtain the endogenous decomposition release and the soil background nutrient holding; S6: Using the soil's baseline nutrient holdings as the initial value, and based on the amount of endogenous decomposition and release, combined with the mineralization and release characteristics of the bio-organic fertilizer to be applied, determine the minimum application amount to ensure that the predicted fertility of the next crop is not lower than the set standard, and then apply it.

[0006] Furthermore, the diurnal variation sequence of soil cellulase activity after harvesting of the target plot was obtained, including: After harvesting in the target plot, soil samples of the stubble residue layer were collected at preset time steps. The activity of soil cellulase in soil samples was measured to obtain the soil cellulase activity value at a single sampling time. Soil cellulase activity values ​​from multiple consecutive single sampling times are arranged in chronological order to form a diurnal variation sequence of soil cellulase activity.

[0007] Furthermore, identifying whether there are activity jump points in the daily activity variation sequence that exceed the daily average background fluctuation range includes: Soil cellulase activity values ​​were selected from the diurnal variation sequence of soil cellulase activity at the first few sampling times after harvest, and the average value was calculated as the daily background value. The standard deviation of soil cellulase activity values ​​at the first few sampling times after harvest was calculated as the background fluctuation range. The sum of the daily average background value and the background fluctuation amplitude is used as the threshold for determining the jump. The soil cellulase activity values ​​at subsequent sampling times in the diurnal variation sequence of soil cellulase activity are compared one by one with the jump threshold. When the soil cellulase activity value at a certain sampling time exceeds the jump threshold for the first time, the time point corresponding to that sampling time is identified as the activity jump point.

[0008] Furthermore, if present, starting from the time corresponding to the activity jump point, the temporal monitoring values ​​of available soil nutrients in the planting area are obtained as the first temporal monitoring values, and the temporal monitoring values ​​of available soil nutrients in the control area without root stubble residue are obtained as the second temporal monitoring values, including: Designate planting areas within the target plots, and preserve the root stubble remaining after harvesting within the planting areas; Within the target plot, a control area with no stubble residue was delineated, and all stubble residue remaining after harvesting in the control area with no stubble residue was removed. Starting from the time corresponding to the activity jump point, soil samples from the planting area and soil samples from the control area without root stubble residue are collected synchronously according to the preset time step. The available nutrient content of soil samples from the planting area was measured, and the available nutrient content of soil at each sampling time was arranged in chronological order to obtain the first time-series monitoring value. The available nutrient content of soil samples from the control area without stubble residue was measured, and the available nutrient content of soil at each sampling time was arranged in chronological order to obtain the second time-series monitoring value.

[0009] Further, the ratio of the initial upward slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value is calculated. If the ratio is within a preset range, the difference curve is obtained by subtracting the nutrient value at the corresponding time of the second time-series monitoring value from the first time-series monitoring value. Based on the difference curve, the corrected initial upward slope and the time point corresponding to the curvature maximum are obtained, including: Linear fitting is performed on the nutrient values ​​at the first few sampling times from the time corresponding to the activity jump point in the first time series monitoring value, and the slope of the fitted line is used as the initial rising slope of the first time series monitoring value. Linear fitting was performed on the nutrient values ​​at the first few sampling times from the time corresponding to the activity jump point in the second time series monitoring value, and the slope of the fitted line was used as the background mineralization slope of the second time series monitoring value. Calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value; Determine whether the ratio of the initial upward slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value is within a preset range; If it is within the preset range, the nutrient value at each sampling time in the first time series monitoring value is subtracted from the nutrient value at the corresponding sampling time in the second time series monitoring value to obtain the difference curve; Calculate the first-order difference sequence between each adjacent sampling time on the difference curve, identify the starting point of the segment in the first-order difference sequence where multiple consecutive negative values ​​first appear, and take the starting point of this segment as the critical sampling time for corrosion decay. The difference curve segment between the time corresponding to the activity jump point and the critical sampling time of corrosion decay is taken as the corrosion-dominant segment. The nutrient values ​​at the initial sampling times in the decomposition-dominant segment are linearly fitted, and the slope of the fitted line is used as the corrected initial upward slope. The curvature of the corrosion-dominant segment is calculated to obtain the sampling time when the curvature of the corrosion-dominant segment reaches its maximum value. This sampling time is taken as the time point corresponding to the maximum curvature value.

[0010] Furthermore, the first-order difference sequence between adjacent sampling times on the difference curve is calculated, and the starting point of the segment in the first-order difference sequence where multiple consecutive negative values ​​first appear is identified. This starting point of the segment is taken as the critical sampling time for decomposition decay. This includes: arranging the nutrient values ​​of each sampling time on the difference curve in chronological order; calculating the difference between the nutrient value of the next sampling time and the nutrient value of the previous sampling time, arranging the differences of all adjacent sampling times in chronological order to obtain the first-order difference sequence; traversing backward from the starting position of the first-order difference sequence, when traversing to a certain position, checking whether several consecutive difference values ​​from that position are all negative; if several consecutive difference values ​​from that position are all negative, then the sampling time corresponding to that position is identified as the starting point of the segment where multiple consecutive negative values ​​first appear, and is taken as the critical sampling time for decomposition decay.

[0011] Furthermore, by inputting the corrected initial upward slope and the time points corresponding to the curvature maxima into the source apportionment relation, the endogenous decomposition release and soil background nutrient holdings are obtained, including: Obtain the pre-calibrated, corrected initial upward slope and the intrinsic decay release as the source resolution relationship; Input the corrected initial rising slope into the source analysis relation to obtain the intrinsic corruption exponent; Obtain the available soil nutrient content at the time point corresponding to the maximum curvature value in the control area without root stubble residue; The soil available nutrient content at the time point corresponding to the maximum curvature in the control area without root stubble residue was used as the soil background nutrient holdings.

[0012] Furthermore, the correspondence between the pre-calibrated corrected initial upward slope and the amount of endogenous humic release is obtained as a source apportionment relationship, including: selecting a calibration plot with the same soil type as the target plot and setting multiple calibration treatments with different root residue amounts; performing steps S1 to S4 on each calibration treatment to obtain the corrected initial upward slope corresponding to each calibration treatment; measuring the total net increase in available soil nutrients for each calibration treatment during the calibration period as the amount of endogenous humic release corresponding to each calibration treatment; and correlating the corrected initial upward slope corresponding to each calibration treatment with the amount of endogenous humic release corresponding to each calibration treatment to establish a correspondence between the corrected initial upward slope and the amount of endogenous humic release, which is used as a source apportionment relationship.

[0013] Furthermore, using the soil's baseline nutrient holdings as an initial value, and based on the endogenous decomposition release rate, combined with the mineralization and release characteristics of the bio-organic fertilizer to be applied, the minimum application rate is determined and applied to ensure that the predicted fertility of the next crop is not lower than the set standard. This includes: The target fertilizer requirement for obtaining the next crop of forage is used as the standard for setting; The soil's baseline nutrient levels are used as the initial fertilizer supply; the amount released from endogenous humification is used as the endogenous fertilizer supply. The exogenous fertilizer requirement is obtained by subtracting the initial fertilizer supply from the target fertilizer requirement and then subtracting the endogenous fertilizer supply. The seasonal mineralization rate of the bedding bio-organic fertilizer was used as a mineralization release characteristic. The minimum application rate is obtained by dividing the required amount of exogenous fertilizer by the current season's mineralization rate of the bio-organic fertilizer in the bedding material. Apply at the minimum recommended dosage.

[0014] On the other hand, the present invention provides a pasture planting organic fertilizer application system based on fertility prediction, comprising: The sequence acquisition module is used to acquire the diurnal variation sequence of soil cellulase activity after harvesting of the target plot; The jump identification module is used to identify whether there are active jump points in the daily change sequence that exceed the range of daily average background fluctuations; The monitoring and acquisition module is used to acquire, if present, the soil effective nutrient time series monitoring value of the planting area as the first time series monitoring value, starting from the time corresponding to the active jump point, and acquire the soil effective nutrient time series monitoring value of the control area without root stubble residue as the second time series monitoring value. The correction processing module is used to calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, the difference curve is obtained by subtracting the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value. Based on the difference curve, the corrected initial rising slope and the time point corresponding to the curvature maximum value are obtained. The source apportionment module is used to input the corrected initial upward slope and the time point corresponding to the curvature maximum into the source apportionment relation to obtain the endogenous humic release and the soil background nutrient holdings; The application determination module is used to determine the minimum application amount that will ensure the predicted fertility of the next crop is not lower than the set standard, based on the soil's background nutrient holdings as the initial value, the amount of endogenous humification release, and the mineralization release characteristics of the bio-organic fertilizer to be applied.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By identifying the diurnal variation sequence of soil cellulase activity and extracting the activity jump point, the initiation of rapid decomposition of stubble was transformed from manual estimation based on harvest time to automatic determination based on enzyme activity biochemical indicators. This synchronizes the starting time of subsequent soil available nutrient monitoring with the actual process of stubble decomposition and release, eliminating the monitoring window misalignment caused by fixed start time. By setting up a control area with no stubble residue and simultaneously collecting the first and second time-series monitoring values, the nutrient dynamics of the planting area under the combined effects of stubble decomposition and background mineralization were obtained, as well as the pure signal of the control area contributed only by background mineralization. This provides a physical reference for solving the source confusion problem.

[0016] 2. By calculating the ratio of the initial upward slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value, difference calculation is performed only when the ratio is within a preset range to generate a difference curve. This achieves graded judgment and selective stripping of the source confusion degree. The operation of identifying the critical sampling time of decomposition decay and limiting the decomposition-dominant segment on the difference curve based on the first-order difference sequence eliminates the tailing interference of slow decomposition components in the later stage of decomposition on the slope and curvature characteristic parameters. This makes the time points corresponding to the corrected initial upward slope and the maximum curvature more accurately reflect the characteristics of rapid decomposition release of the root stubble. After obtaining the endogenous decomposition release and the soil background nutrient holding through source apportionment relationship, the two, together with the current season mineralization rate of the bedding bio-organic fertilizer, participate in the calculation of the exogenous fertilizer requirement. The final determined minimum application amount completely deducts the dual contribution of soil background nutrient supply and endogenous decomposition release, avoiding over-application caused by repeated inclusion of endogenous nutrients. This realizes the precise application of bedding bio-organic fertilizer on demand in multi-crop forage planting systems. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for applying organic fertilizer to forage grass cultivation based on fertility prediction, according to the present invention. Figure 2 This is a schematic diagram of the structure of an organic fertilizer application system for forage planting based on fertility prediction according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1 This invention provides a method for applying organic fertilizer to forage crops based on fertility prediction, comprising: S1: Obtain the diurnal variation sequence of soil cellulase activity after harvesting in the target plot; S2: Identify whether there are activity jump points in the daily activity variation sequence that exceed the range of daily average background fluctuations; S3: If it exists, then starting from the time corresponding to the active jump point, obtain the soil effective nutrient time-series monitoring value of the planting area as the first time-series monitoring value, and obtain the soil effective nutrient time-series monitoring value of the control area without root stubble residue as the second time-series monitoring value. S4: Calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, subtract the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value to obtain the difference curve. Based on the difference curve, obtain the corrected initial rising slope and the time point corresponding to the maximum curvature. S5: Input the corrected initial upward slope and the time point corresponding to the curvature maximum value into the source analysis relation to obtain the endogenous decomposition release and the soil background nutrient holding; S6: Using the soil's baseline nutrient holdings as the initial value, and based on the amount of endogenous decomposition and release, combined with the mineralization and release characteristics of the bio-organic fertilizer to be applied, determine the minimum application amount to ensure that the predicted fertility of the next crop is not lower than the set standard, and then apply it.

[0020] In the specific implementation of S1, after the target plot is harvested, soil samples from the stubble residue layer are collected at preset time steps. After the current crop of forage is harvested, soil samples are collected every preset time step, starting from the moment the harvest is completed. The stubble residue layer refers to the soil layer consisting of the forage stubble and attached soil after harvesting, and the sampling depth is consistent with the distribution depth of the stubble in the soil. The preset time step is determined empirically based on the stubble decomposition rate of the forage variety. Forage varieties with faster decomposition rates correspond to shorter preset time steps. For example, the preset time step for grass forage is set to 4 to 6 hours, and for leguminous forage is set to 6 to 8 hours. For each sampling, for example, three sampling points are randomly selected within the planting area. After scraping away the exposed fallen leaves, soil samples from the stubble residue layer are collected. The soil samples from each sampling point collected in the same session are mixed evenly to serve as the stubble residue layer soil sample corresponding to that single sampling time.

[0021] Soil cellulase activity was determined in soil samples to obtain the activity value at a single sampling time. For each collected root stubble soil sample, the 3,5-dinitrosalicylic acid colorimetric method was used to determine soil cellulase activity. Specifically, a sodium carboxymethyl cellulose solution was added to the root stubble soil sample as a substrate and incubated for 24 hours under buffer conditions, such as 37°C and pH 5.5. After incubation, 3,5-dinitrosalicylic acid was added to terminate the reaction. The resulting mixture was then subjected to colorimetric analysis, and the absorbance value at 540 nm was read. The absorbance value was converted to glucose production based on the glucose standard curve. Soil cellulase activity was defined as the number of milligrams of glucose produced per gram of dried soil from the hydrolysis of sodium carboxymethyl cellulose within 24 hours, expressed as milligrams of glucose per gram of soil per 24 hours. Each root stubble soil sample, after the above determination, yielded a soil cellulase activity value at a single sampling time, corresponding to the sampling time.

[0022] Soil cellulase activity values ​​from multiple consecutive single sampling times are arranged chronologically to form a diurnal variation sequence of soil cellulase activity. The soil cellulase activity values ​​from each single sampling time are then arranged chronologically from earliest to latest, forming an ordered data column, which is the diurnal variation sequence of soil cellulase activity. In this sequence, the x-axis represents the sampling time, and the y-axis represents the corresponding soil cellulase activity value. After constructing the diurnal variation sequence, subsequent steps involve identifying any activity jumps that exceed the daily background fluctuation range.

[0023] The diurnal variation sequence of soil cellulase activity is used as the criterion for determining whether rapid decomposition of stubble has begun. The principle is as follows: after forage harvesting, stubble residue begins to decompose under the action of soil microorganisms. During decomposition, microorganisms attach to the stubble and secrete cellulase to break down the cellulose in the stubble cell walls, leading to a significant increase in soil cellulase activity in the stubble residue layer. This increase in activity occurs earlier than the substantial increase in available soil nutrients. By continuously monitoring soil cellulase activity and constructing a diurnal variation sequence, the initiation signal of rapid stubble decomposition can be captured before the large-scale release of decomposition products. This provides an accurate starting point for subsequent time-series monitoring of available soil nutrients, starting from the time corresponding to the activity jump point. Compared with existing technologies that use a fixed time based on the number of days after harvest, using the activity jump point in the diurnal variation sequence of soil cellulase activity as the starting point avoids the problem of the monitoring window being out of sync with the decomposition process due to factors such as soil temperature and humidity and the degree of lignification of the stubble. This makes the subsequent quantitative analysis of nutrients released during stubble decomposition more accurate.

[0024] In the specific implementation of S2, soil cellulase activity values ​​at several initial sampling times after harvest are selected from the diurnal variation sequence of soil cellulase activity, and the average value is calculated as the daily background value. The diurnal variation sequence of soil cellulase activity is generated by the step of obtaining the diurnal variation sequence of soil cellulase activity after harvest in the target plot. This sequence contains soil cellulase activity values ​​at multiple single sampling times arranged in chronological order of sampling time. The initial sampling times after harvest refer to all single sampling times included within the initial period of time from the completion of harvest. This period is set as the length of the stable period before the substantial initiation of stubble decomposition. The length of the stable period is determined according to the forage variety and soil temperature and humidity conditions. Taking gramineous forage under the condition of a daily average soil temperature of 15°C to 25°C as an example, the length of the stable period is taken as, for example, 12 hours to 24 hours after harvest. During this period, the soil cellulase activity value only reflects the inherent background enzyme activity in the soil and has not yet been disturbed by the rapid decomposition process of stubble. The daily background value is obtained by summing all soil cellulase activity values ​​at the initial sampling times after harvest and dividing by the number of soil cellulase activity values ​​included in the summation. The unit of the daily background value is the same as that of the soil cellulase activity value, which is milligrams of glucose per gram of soil per 24 hours. The daily background value represents the normal level of soil cellulase activity in the root stubble residue layer of the target plot when rapid root stubble decomposition has not occurred.

[0025] The standard deviation of soil cellulase activity values ​​at the initial sampling times after harvest is used as the background fluctuation range. The standard deviation is calculated as follows: first, calculate the average soil cellulase activity value at the initial sampling times after harvest, i.e., the daily average background value; then, calculate the square of the difference between the soil cellulase activity value at each sampling time and the daily average background value. Sum all the squares, divide by the number of soil cellulase activity values ​​used in the calculation minus 1, and then take the square root of the quotient to obtain the standard deviation. This standard deviation is the background fluctuation range. The unit of the background fluctuation range is the same as that of the soil cellulase activity value: milligrams of glucose per gram of soil per 24 hours. The background fluctuation range characterizes the normal fluctuation degree of soil cellulase activity around the daily average background value before the rapid decomposition of stubble begins. The fluctuation originates from the combined effects of soil microenvironment differences, the randomness of microbial activity, and sampling and measurement errors.

[0026] The sum of the daily average background value and the background fluctuation amplitude is used as the threshold for determining the jump. The jump threshold is set as the direct sum of the daily average background value and the background fluctuation amplitude, with the unit consistent with that of soil cellulase activity: milligrams of glucose per gram of soil per 24 hours. The basis for setting the jump threshold is that when the soil cellulase activity value at a certain sampling time exceeds the sum of the daily average background value and the background fluctuation amplitude, it means that the soil cellulase activity value has exceeded the upper limit that can be explained by normal random fluctuations. It can be determined that this exceedance is not caused by sampling errors or random fluctuations in the microenvironment, but rather by a substantial increase in soil cellulase activity caused by the rapid decomposition of stubble. The value of the jump threshold depends on two parameters: the daily average background value and the background fluctuation amplitude. Both parameters are statistically derived from the soil cellulase activity values ​​measured in the first few sampling times after harvest, which are actually measured in the daily variation sequence of soil cellulase activity. This allows the jump threshold to adapt to the actual soil conditions of different plots and different cropping times, without relying on external empirical parameters or historical databases.

[0027] The soil cellulase activity values ​​at subsequent sampling times in the diurnal variation sequence of soil cellulase activity are compared one by one with the jump threshold. When the soil cellulase activity value at a certain sampling time first exceeds the jump threshold, the time point corresponding to that sampling time is identified as the activity jump point. Subsequent sampling times refer to all sampling times in the diurnal variation sequence of soil cellulase activity, starting from the last sampling time of the first few sampling times after harvest.

[0028] The process iterates through each subsequent sampling time, comparing the soil cellulase activity value at each sampling time with the jump threshold. The comparison order strictly follows the chronological order of the sampling times. When a sampling time is reached, if the soil cellulase activity value at that time is greater than the jump threshold, the sampling time meets the condition for the first exceedance, the iteration stops, and the time point corresponding to that sampling time is identified as the activity jump point. If the soil cellulase activity value at that sampling time is less than or equal to the jump threshold, the comparison continues to the next sampling time.

[0029] If, after traversing all subsequent sampling times, no soil cellulase activity value exceeds the threshold for a jump, it is determined that there are no activity jump points exceeding the daily background fluctuation range in the diurnal variation sequence of soil cellulase activity. At this point, rapid decomposition of stubble has not yet substantially begun. The subsequent step of obtaining soil available nutrient time-series monitoring values ​​is not initiated. Instead, soil samples from the stubble residue layer are collected at preset time steps, and the diurnal variation sequence of soil cellulase activity is updated. The step of identifying activity jump points exceeding the daily background fluctuation range in the diurnal variation sequence is then repeated. The time point corresponding to the activity jump point is the time marker for the substantial initiation of rapid decomposition of stubble.

[0030] In the initial short period after harvest, the stubble surface is not yet fully colonized by microorganisms, and cellulase activity remains at a low background level. The soil cellulase activity values ​​measured at this time can be used to estimate the normal fluctuation range of background enzyme activity. As time progresses post-harvest, microorganisms begin to multiply and attach to the stubble, resulting in a detectable surge in cellulase activity. Once this surge exceeds the normal fluctuation range, it can be determined that rapid stubble decomposition has substantially begun. Compared to methods such as directly observing stubble color changes or measuring stubble weight loss, soil cellulase activity responds more sensitively to the initiation of rapid stubble decomposition and earlier. Using the activity surge point as the starting point for subsequent monitoring of available soil nutrients ensures that the acquisition window for soil available nutrient time-series monitoring values ​​is highly synchronized with the actual stubble decomposition and release process, eliminating monitoring lag or overshoot caused by fixed starting times.

[0031] In the specific implementation of S3, planting areas are demarcated within the target plot, preserving the stubble remaining after harvest. The target plot refers to a plot where the current forage harvest has been completed. A representative area within the target plot is selected as the planting area. The area of ​​the planting area is determined based on the sampling volume and replication number required for subsequent soil nutrient monitoring, and is set to, for example, 10 to 20 square meters. All naturally remaining stubble is preserved within the planting area; no stubble removal is performed, and the stubble residue state within the planting area remains consistent with other harvested areas within the target plot. The boundaries of the planting area are marked on the surface using bamboo sticks or rulers, and the markings must not affect the soil environment or the stubble decomposition process within the planting area.

[0032] Within the target plot, a stubble-free control area was demarcated, and all stubble remaining after harvest was removed from this area. An area within the target plot with the same soil type, terrain conditions, and uniform growth of the previous forage crop as the stubble-free control area was selected. The distance between the stubble-free control area and the planting area was set at, for example, 2 to 5 meters, to ensure that the soil background and microclimate conditions were the same but did not interfere with each other. The area of ​​the stubble-free control area was the same as the area of ​​the planting area to ensure the statistical validity of subsequent sampling and data comparison. Within the stubble-free control area, all remaining stubble was manually removed one by one, taking care not to disturb the surrounding soil structure. After removal, the soil surface of the stubble-free control area was inspected to confirm that no visible stubble fragments remained. After removing the stubble, a breathable and water-permeable isolation layer, made of nylon mesh or gauze, was covered on the surface of the stubble-free control area. The purpose of the isolation layer is to prevent fallen leaves from the planting area or other areas from being blown into the stubble-free control area by the wind, thus preventing additional organic matter disturbance, while not affecting soil gas exchange and water infiltration in the stubble-free control area. No organic fertilizer is applied to the stubble-free control area, and all other management practices are completely consistent with those in the planting area.

[0033] Starting from the time corresponding to the activity jump point, soil samples from the planting area and the control area without root stubble residue are collected synchronously at a preset time step. The time corresponding to the activity jump point is determined by comparing the soil cellulase activity values ​​at subsequent sampling times in the diurnal variation sequence of soil cellulase activity with the jump threshold. When the soil cellulase activity value at a certain sampling time first exceeds the jump threshold, the time point corresponding to that sampling time is identified as the activity jump point. Starting from the time point corresponding to the activity jump point, synchronous soil sampling begins in the planting area and the control area without root stubble residue. The preset time step is determined based on the rate characteristics of nutrient release from root stubble decomposition. The preset time step is set to, for example, 6 to 12 hours, to ensure that a sufficient density of data points can be collected within a complete decomposition cycle for calculating the initial upward slope of the first time-series monitoring values ​​and generating the difference curve.

[0034] Synchronous sampling means that at each preset time step, soil samples are collected sequentially from the planting area and the stubble-free control area, with the time interval between the two samplings controlled, for example, within 15 minutes, to ensure that the soil samples from both areas represent the same time point. The soil samples from the planting area are collected by randomly selecting, for example, three sampling points within the planting area, scraping away visible fallen leaves, and collecting soil from the stubble layer. The soil from the three sampling points is then mixed thoroughly to form a single soil sample from the planting area. Similarly, the soil samples from the stubble-free control area are collected by randomly selecting, for example, three sampling points within the stubble-free control area, avoiding the isolation layer, and collecting soil from the depth of the stubble layer. The soil from the three sampling points is then mixed thoroughly to form a single soil sample from the stubble-free control area. Each soil sample from the planting area and the stubble-free control area is individually numbered, and the sampling time is recorded.

[0035] The available nutrient content of soil samples from the planting area was determined, and the available nutrient content at each sampling time was arranged in chronological order to obtain the first time-series monitoring value. Available nutrient content refers to the content of mineral nutrients in the soil that can be absorbed and utilized by the current season's forage grass. Available nutrient content includes at least alkaline nitrogen, available phosphorus, and available potassium. Taking the determination of alkaline nitrogen content as an example: soil samples from the planting area were taken and determined using the alkaline diffusion method. The soil sample was placed in the outer chamber of a diffusion dish, and sodium hydroxide solution and a reducing agent were added. A boric acid indicator mixture was added to the inner chamber of the diffusion dish. After diffusion under constant temperature conditions, the solution in the inner chamber was removed and titrated with standard acid. The alkaline nitrogen content was calculated from the amount of standard acid consumed, with units of milligrams per kilogram. Available phosphorus content was determined using the sodium bicarbonate extraction-molybdenum antimony colorimetric method, and available potassium content was determined using the ammonium acetate extraction-flame photometric method. The soil available nutrient content at each sampling time was arranged chronologically to obtain an ordered data series, which is the first time-series monitoring value. The first time-series monitoring value reflects the dynamic changes in soil available nutrients in the planting area under the combined effects of root stubble decomposition and soil background mineralization.

[0036] The available nutrient content of soil samples from the stubble-free control area was determined, and the available nutrient content at each sampling time was arranged chronologically to obtain the second time-series monitoring values. Using the same method as for the planting area, the available nutrient content of soil samples from the stubble-free control area was determined at each sampling time. The available nutrient content at each sampling time and the available nutrient content at each sampling time were arranged chronologically to obtain an ordered data series, which constitutes the second time-series monitoring values. The second time-series monitoring values ​​reflect the dynamic changes in available nutrients in the stubble-free control area under conditions of only background soil mineralization and no interference from stubble decomposition, providing a benchmark for subsequently subtracting the background mineralization contribution from the first time-series monitoring values.

[0037] The necessity of setting up a stubble-free control area lies in the fact that conventional methods only monitor changes in available soil nutrients in the planting area, failing to distinguish whether the increase in available soil nutrients originates from root stubble decomposition or the continuous mineralization of existing soil organic matter. By simultaneously setting up a stubble-free control area within the target plot, a pure soil background mineralization signal can be obtained under the same soil type, climate conditions, and sampling time step. This provides a physically meaningful benchmark for subsequent differential calculations to separate endogenous decomposition release, rather than relying on mathematical model assumptions for source separation. It also resolves the technical problem of confusion between previous crop residue decomposition release and soil background mineralization, ensuring the accuracy of the ratio of the initial upward slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value, as well as the accuracy of the time points corresponding to the corrected initial upward slope and the curvature maxima obtained based on the difference curve.

[0038] In the specific implementation of S4, the nutrient values ​​of the initial few sampling times from the time corresponding to the activity jump point in the first time-series monitoring value are linearly fitted, and the slope of the fitted straight line is used as the initial rising slope of the first time-series monitoring value. The first time-series monitoring value is obtained by measuring the available soil nutrient content of soil samples from the planting area and arranging the available soil nutrient content of each sampling time in chronological order. The time corresponding to the activity jump point is the time point corresponding to the activity jump point identified when the soil cellulase activity value of subsequent sampling times in the diurnal variation sequence of soil cellulase activity is compared one by one with the jump judgment threshold. The initial few sampling times refer to all sampling times included within a period of time starting from the time point corresponding to the activity jump point. This period of time is set as the length of the linear release stage in the initial stage of rapid decomposition of stubble.

[0039] The length of the linear release phase is determined based on the forage variety and soil temperature and humidity conditions. Taking gramineous forage under a daily average soil temperature of 15°C to 25°C as an example, the linear release phase length is taken as 24 to 48 hours after the time point corresponding to the activity jump point. During this period, the rate of nutrient release from root stubble decomposition remains approximately constant. Linear fitting uses the least squares method. The sampling time values ​​corresponding to each sampling time in the initial sampling time and the nutrient values ​​at that sampling time in the first time-series monitoring values ​​are used as two-dimensional coordinate points. A straight line is found that minimizes the sum of the squares of the vertical distances from all two-dimensional coordinate points to this line. The slope of the fitted line is the initial upward slope of the first time-series monitoring values. The unit of the initial upward slope of the first time-series monitoring values ​​is nutrient content per unit time. The initial upward slope of the first time-series monitoring values ​​characterizes the initial growth rate of available soil nutrients in the planting area under the combined effects of root stubble decomposition and soil background mineralization.

[0040] Linear fitting was performed on the nutrient values ​​at the initial sampling times corresponding to the activity jump point in the second time-series monitoring values. The slope of the fitted line was used as the background mineralization slope of the second time-series monitoring values. The second time-series monitoring values ​​were obtained by measuring the available soil nutrient content of soil samples from the stubble-free control area and arranging the available soil nutrient content at each sampling time in chronological order. The time range of the initial sampling times used for linear fitting of the second time-series monitoring values ​​was exactly the same as that used for linear fitting of the first time-series monitoring values, both calculated from the time point corresponding to the activity jump point with the same linear release phase length. The least squares method was also used for linear fitting, and the slope of the fitted line was taken as the background mineralization slope of the second time-series monitoring values. The unit of the background mineralization slope of the second time-series monitoring values ​​was consistent with the unit of the initial rising slope of the first time-series monitoring values. The background mineralization slope of the second time-series monitoring values ​​characterizes the growth rate of available soil nutrients in the stubble-free control area under only soil background mineralization. Since all roots and stubble had been removed from the control area without root stubble residue, and there was no root stubble decomposition and release process, the background mineralization slope of the second time series monitoring value reflects the contribution rate of the continuous mineralization of the original organic matter in the soil to the available nutrients in the soil in the control area without root stubble residue.

[0041] Calculate the ratio of the initial rising slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value. Using the initial rising slope of the first time-series monitoring value as the dividend and the background mineralization slope of the second time-series monitoring value as the divisor, the two are divided to obtain a dimensionless ratio. This ratio reflects the multiple relationship between the initial rising slope of the first time-series monitoring value and the background mineralization slope of the second time-series monitoring value.

[0042] The calculation determines whether the ratio of the initial upward slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value falls within a preset interval. The preset interval is a closed interval predetermined based on the ratio range corresponding to different contribution levels of root stubble decomposition. The lower limit of the preset interval is set, for example, 1.2, and the upper limit is set, for example, 3.0. The preset interval is set based on the following: when the ratio is less than the lower limit, it indicates that the difference between the initial upward slope of the first time-series monitoring value and the background mineralization slope of the second time-series monitoring value is not significant, and the contribution of root stubble decomposition to endogenous nutrients is weak. Therefore, it is not necessary to separately separate the amount of endogenous decomposition and the soil background nutrient holdings; the second time-series monitoring value can be directly used as the soil background mineralization benchmark for subsequent fertilization decisions. When the ratio is greater than the upper limit of the preset interval, it indicates that root stubble decomposition is absolutely dominant, and the contribution of soil background mineralization is relatively negligible. The first time-series monitoring value can be directly approximated as a signal of endogenous decomposition for subsequent processing. When the ratio is within the preset range, it indicates that the contribution of root stubble decomposition is comparable to that of soil background mineralization, and there is a risk of source confusion. The two must be separated by difference calculation.

[0043] If the ratio is within a preset range, the nutrient values ​​at each sampling time in the first time series monitoring value are subtracted from the nutrient values ​​at the corresponding sampling time in the second time series monitoring value to obtain a difference curve. When the ratio of the initial upward slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value is within a preset range, the difference operation is performed. The difference operation involves subtracting the nutrient value at each sampling time in the first time series monitoring value as the minuend and the nutrient value at the same sampling time in the second time series monitoring value as the subtrahend, obtaining a series of difference nutrient values ​​corresponding to the sampling times. The sampling times and the corresponding difference nutrient values ​​are arranged in chronological order to form an ordered data series, which is the difference curve. The difference curve eliminates the continuous contribution of soil background mineralization to soil available nutrients, retaining only the dynamic change of the net contribution of root stubble decomposition to soil available nutrients over time.

[0044] Calculate the first-order difference sequence between adjacent sampling times on the difference curve, identify the starting point of the segment in the first-order difference sequence where multiple consecutive negative values ​​first appear, and use this starting point as the critical sampling time for decomposition decay. Arrange the difference nutrient values ​​at each sampling time on the difference curve in chronological order. Starting from the second sampling time on the difference curve, calculate the difference nutrient value at each sampling time by subtracting the difference nutrient value at the previous sampling time. Each difference corresponds to a sampling time interval. Arrange the differences between all adjacent sampling times in chronological order to obtain the first-order difference sequence. Each value in the first-order difference sequence represents the net change in difference nutrient value within the corresponding sampling time interval. A positive value indicates that the difference nutrient value is still increasing, meaning that root stubble decomposition is still in the net release stage, while a negative value indicates that the difference nutrient value has begun to decrease, meaning that root stubble decomposition has entered the net decay stage. The process iterates sequentially from the starting position of the first-order difference sequence. When a certain position is reached, it checks whether three consecutive difference values ​​from that position are all negative. If, for example, three consecutive difference values ​​are negative from that position, the nutrient difference value is determined to have entered a continuous downward trend, rather than a random fluctuation from a single sampling. The sampling time corresponding to this position is identified as the starting point of the segment where multiple consecutive negative values ​​first appear, and is used as the critical sampling time for decomposition attenuation. The number of consecutive values ​​is set based on the following criteria: too small a number may easily misjudge sampling errors or instantaneous fluctuations as trend changes, while too large a number will lead to a lag in the identification of trend changes. A number of 3 is an empirical value that strikes a balance between sensitivity and robustness. The critical sampling time for decomposition attenuation marks the basic end of the rapid decomposition stage of stubble, after which the net contribution of stubble decomposition to the available nutrients in the soil begins to decrease continuously.

[0045] The difference curve segment between the time corresponding to the activity jump point and the critical sampling time of decomposition attenuation is taken as the dominant segment of decomposition. The time point corresponding to the activity jump point marks the substantial start of rapid decomposition of stubble, and the critical sampling time of decomposition attenuation marks the basic end of rapid decomposition of stubble. From the time point corresponding to the activity jump point to the critical sampling time of decomposition attenuation, the decomposition of stubble plays a dominant role in the change of available nutrients in the soil, and the difference in nutrient values ​​during this period is basically unaffected by the slow-decomposing components in the later stages of decomposition. The difference curve corresponding to this period is extracted as the dominant segment of decomposition.

[0046] Nutrient values ​​at the initial sampling times during the predominant decomposition stage were linearly fitted, and the slope of the fitted line was used as the corrected initial upward slope. The selection range of the initial sampling times during the predominant decomposition stage was consistent with the length of the linear release phase. The difference in nutrient values ​​at the initial sampling times during the predominant decomposition stage was linearly fitted using the least squares method, and the slope of the fitted line was used as the corrected initial upward slope. The corrected initial upward slope characterizes the pure initial upward rate of nutrient release from rapid root decomposition after eliminating the contribution of soil background mineralization and interference from slowly decomposed components in the later stages of decomposition.

[0047] The curvature of the dominant corrosion segment is calculated, and the sampling time when the curvature reaches its maximum value is obtained. This sampling time is taken as the time point corresponding to the curvature maximum. The curvature is calculated as: K(t) = |S''(t)| / (1+(S'(t))) 2 ) (3 / 2) Where K(t) represents the curvature value at sampling time t, S'(t) represents the first derivative value at sampling time t on the decomposition-dominant segment, S''(t) represents the second derivative value at sampling time t on the decomposition-dominant segment, and |·| represents the absolute value operation. The first and second derivatives are calculated point-by-point using numerical differentiation. The first derivative value is the difference between the difference in nutrient values ​​at sampling time t and the difference in nutrient values ​​at the previous sampling time, divided by a preset time step. The second derivative value is the difference between the first derivative value at sampling time t and the first derivative value at the previous sampling time, divided by a preset time step. After calculating the curvature values ​​at all sampling times on the decomposition-dominant segment, the magnitudes of the curvature values ​​at all sampling times are compared. The sampling time at which the curvature value reaches its maximum is identified, and this sampling time is taken as the time point corresponding to the curvature maximum. The time point corresponding to the curvature maximum corresponds to the moment when the rate of nutrient release from root decomposition on the decomposition-dominant segment changes most drastically, and is a key time characteristic parameter characterizing the endogenous decomposition process.

[0048] First-order difference sequences were used to identify the critical sampling time for decomposition decay and to define the dominant decomposition segment. Root stubble decomposition release does not maintain a constant rate. In the early stages of rapid decomposition, the nutrient release rate is nearly linear. As easily decomposable components are gradually depleted, the nutrient release rate begins to decline, and the difference curve becomes curved. Conventional methods fit the entire difference curve, failing to consider the tailing interference of slowly decomposed components in the later stages of decomposition, resulting in the fitted slope and curvature characteristic values ​​deviating from the pure rapid decomposition signal. First-order difference sequences were used to identify the turning point from net growth to net decay in decomposition release. The segment of the difference curve from the time point corresponding to the activity jump point to the critical sampling time for decomposition decay was taken as the dominant decomposition segment. This segmented the decomposition process in stages from a temporal perspective, excluding the ambiguous signal of slowly decomposed components in the later stages of decomposition from the analysis window. This allows the corrected initial rising slope and the time point corresponding to the maximum curvature to more accurately reflect the true characteristics of rapid root stubble decomposition release, providing higher-quality input parameters for subsequent source-analysis relationships.

[0049] In the specific implementation of S5, the correspondence between the pre-calibrated corrected initial upward slope and the endogenous humic release is obtained as the source apportionment relationship. The source apportionment relationship is a quantitative mapping relationship established in advance through field calibration experiments, mapping the corrected initial upward slope to the endogenous humic release. The pre-calibration process of the source apportionment relationship is as follows: Calibration plots with the same soil type as the target plot are selected. The soil type of the calibration plots is determined based on soil texture, soil organic matter content, and soil pH to ensure that the soil background conditions of the calibration plots are consistent with those of the target plot. Multiple calibration treatments with different root stubble residues are set up within the calibration plots. For example, a complete removal treatment with zero root stubble residue, a halving treatment with half the natural residue after harvest, a standard treatment with the same root stubble residue as the natural residue after harvest, and a doubling treatment with twice the natural residue after harvest. The stubble residue refers to the weight of dry matter remaining in the soil after harvesting. The stubble residue of each calibration treatment is different by manually adjusting the number of stubble after harvesting. Except for the stubble residue, all other conditions are kept consistent between the calibration treatments.

[0050] For each calibration treatment, the following steps were performed: obtaining the diurnal variation sequence of soil cellulase activity after harvest in the target plot; obtaining the time point corresponding to the corrected initial upward slope and the maximum curvature based on the difference curve; specifically, obtaining the diurnal variation sequence of soil cellulase activity after harvest in the target plot; identifying whether there are activity jump points in the diurnal variation sequence that exceed the daily average background fluctuation range; if so, obtaining the soil available nutrient time series monitoring value of the planting area as the first time series monitoring value from the time corresponding to the activity jump point; obtaining the soil available nutrient time series monitoring value of the control area without stubble residue as the second time series monitoring value; calculating the ratio of the initial upward slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value; if the ratio is within a preset range, subtracting the nutrient value at the time corresponding to the second time series monitoring value from the first time series monitoring value to obtain the difference curve; and obtaining the time point corresponding to the corrected initial upward slope and the maximum curvature based on the difference curve to obtain the corrected initial upward slope for each calibration treatment. The total net increase of soil available nutrients for each calibration treatment during the calibration period was measured. The calibration period was set from the time point corresponding to the activity jump point until the complete decomposition of root stubble. The complete decomposition of root stubble was judged by the flattening of the difference curve. The flattening of the difference curve was judged by the fact that the change in the difference nutrient value at multiple consecutive sampling times on the difference curve was less than 5% of the maximum difference nutrient value on the difference curve, and this state lasted for more than 24 hours. The calibration period was usually 7 to 14 days. At the beginning of the calibration period, the initial content of available soil nutrients for each calibration treatment was measured, and at the end of the calibration period, the final content of available soil nutrients for each calibration treatment was measured. The net increase of available soil nutrients for each calibration treatment was obtained by subtracting the initial content of available soil nutrients from the final content of available soil nutrients. Since no organic fertilizer was applied to any of the calibration treatments in the calibration plot, the net increase of available soil nutrients for each calibration treatment came entirely from the decomposition release of the root stubble residue. Therefore, the net increase of available soil nutrients for each calibration treatment was the endogenous decomposition release amount corresponding to each calibration treatment. The corrected initial upward slope corresponding to each calibration treatment is used as the independent variable, and the endogenous decay expulsion corresponding to each calibration treatment is used as the dependent variable. The correspondence between the corrected initial upward slope and the endogenous decay expulsion is established. The correspondence between the corrected initial upward slope and the endogenous decay expulsion is stored in the form of an interpolation table or a fitted curve. This correspondence is the source resolution relationship.

[0051] The corrected initial upward slope is input into the source apportionment relation to obtain the endogenous humification release. The corrected initial upward slope is obtained by linearly fitting the nutrient values ​​at the initial sampling times during the decomposition-dominant phase. The value of the corrected initial upward slope is input into the pre-calibrated source apportionment relation. If the source apportionment relation is stored in the form of an interpolation table, the two calibration values ​​closest to the input corrected initial upward slope are found in the interpolation table and linearly interpolated to obtain the corresponding endogenous humification release. If the source apportionment relation is stored in the form of a fitted curve, the input corrected initial upward slope is substituted into the function expression of the fitted curve to calculate the corresponding endogenous humification release. The endogenous humification release represents the portion of the total amount of available soil nutrients released by root stubble decomposition that can be absorbed and utilized by the next crop of forage grass during the entire cycle from the time point corresponding to the activity jump point to the complete end of rapid root stubble decomposition. The unit of endogenous humification release is milligrams per kilogram, consistent with the unit of available soil nutrient content. Since the source apportionment relationship is obtained by controlling different root residue amounts on a calibration plot with the same soil type as the target plot, the endogenous humic release obtained by inputting the corrected initial upward slope can reflect the true endogenous humic release level of the target plot under natural root residue conditions.

[0052] The available soil nutrient content of the control area without stubble residue was obtained at the time point corresponding to the curvature maximum. The time point corresponding to the curvature maximum was obtained by calculating the curvature of the decomposition-dominant segment. In the second time-series monitoring data, a sampling time corresponding to the curvature maximum was found, and the available soil nutrient content of the second time-series monitoring data corresponding to this sampling time was extracted. This available soil nutrient content is the available soil nutrient content of the control area without stubble residue at the time point corresponding to the curvature maximum. The time point corresponding to the curvature maximum corresponds to the moment when the rate of nutrient release from root stubble decomposition in the decomposition-dominant segment changes most drastically. The available soil nutrient content of the control area without stubble residue at this moment was chosen as the time point for determining the soil background nutrient holdings because this moment is synchronized with the key turning point in the endogenous decomposition process, and can most accurately reflect the nutrient level contributed by soil background mineralization when the endogenous decomposition rate reaches its peak.

[0053] The soil available nutrient content at the time point corresponding to the maximum curvature in the stubble-free control area was used as the soil background nutrient holding. Soil background nutrient holding is defined as the amount of available soil nutrients that can be absorbed and utilized by the next crop of forage, continuously supplied by background mineralization, at the time point when stubble decomposition reaches its peak rate. Since all stubble in the stubble-free control area was removed, there was no stubble decomposition process; the soil available nutrient content in the stubble-free control area was entirely contributed by the continuous mineralization of the original soil organic matter. Therefore, using the soil available nutrient content at the time point corresponding to the maximum curvature in the stubble-free control area as the soil background nutrient holding has a clear physical meaning. The unit of soil background nutrient holding is consistent with the unit of soil available nutrient content, which is milligrams per kilogram.

[0054] Existing technologies directly use soil nutrient measurements at a fixed point after harvest as the initial state for next crop fertilization decisions. These measurements mix the released but not yet absorbed endogenous humic emissions with the soil's background nutrient holdings, making numerical distinction between the two impossible. By using a measurable kinetic parameter—the corrected initial slope—as a bridge, and leveraging a pre-established source apportionment relationship on similar soils, the endogenous humic emissions can be separately analyzed. Simultaneously, the soil's background nutrient holdings are obtained from a physical baseline—a control area without stubble—at the time point corresponding to the curvature maxima. This achieves dual separation of endogenous humic emissions and soil background nutrient holdings in both time and source dimensions, avoiding fertilization calculation errors caused by source confusion.

[0055] In the specific implementation of S6, the target nutrient requirement for the next crop of forage grass is used as the setting standard. The target nutrient requirement refers to the total amount of available soil nutrients that the next crop of forage grass needs to absorb to reach the expected yield level under normal growth conditions. The target nutrient requirement is determined as follows: based on the variety of the next crop of forage grass and the expected target yield, the nutrient absorption per unit yield of that forage grass variety under normal cultivation conditions in the current planting area is consulted. The nutrient absorption per unit yield is then multiplied by the expected target yield to obtain the target nutrient requirement. For example, when planting ryegrass with an expected target yield of, for example, 3000 kg of fresh grass per acre, the amount of available nitrogen absorbed from the soil for every 1000 kg of fresh grass produced by ryegrass is approximately 3.5 kg. Therefore, the amount of available nitrogen in the target nutrient requirement is approximately 3.5 × 3 = 10.5 kg per acre. The unit of the target fertilizer requirement is consistent with the unit of the soil's available nutrient content. In actual calculations, the target fertilizer requirement is converted to milligrams per kilogram, based on the weight of the topsoil per acre, for example, 150,000 kilograms. The target fertilizer requirement serves as a benchmark for judging whether the predicted fertility for the next crop is sufficient.

[0056] The soil background nutrient holdings are used as the initial fertilization amount. Soil background nutrient holdings are obtained by using the soil available nutrient content at the time point corresponding to the maximum curvature in the control area without stubble residue as the soil background nutrient holdings. Soil background nutrient holdings characterize the amount of available soil nutrients continuously supplied by background mineralization when stubble decomposition reaches its peak rate. The soil background nutrient holdings are directly used as the initial fertilization amount, which is the amount of available nutrients already present in the soil at the start of the next forage growing season that can be absorbed and utilized by forage grasses. The unit of the initial fertilization amount is the same as that of the soil background nutrient holdings, in milligrams per kilogram.

[0057] The amount of nutrients released through endogenous decomposition is taken as the endogenous nutrient supply. This endogenous decomposition release is obtained by inputting the corrected initial ascent slope into the source resolving power relation. The endogenous decomposition release characterizes the portion of the total available soil nutrients released by root stubble decomposition that can be absorbed and utilized by the next crop of forage grass during the entire cycle from the time point corresponding to the activity jump point to the complete end of rapid root stubble decomposition. The endogenous nutrient supply is directly taken as the endogenous nutrient supply, which is the amount of available nutrients that can be absorbed and utilized by forage grass due to the continuous decomposition of root stubble during the next forage grass growing season. The unit of endogenous nutrient supply is the same as that of endogenous decomposition release, which is milligrams per kilogram.

[0058] The exogenous fertilizer requirement is calculated by subtracting the initial fertilizer supply and then the endogenous fertilizer supply from the target fertilizer requirement. The exogenous fertilizer requirement is calculated as: Fex = Ftarget - Ninitial - Nendo; where Fex represents the exogenous fertilizer requirement, Ftarget represents the target fertilizer requirement, Ninitial represents the initial fertilizer supply, and Nendo represents the endogenous fertilizer supply. The units for Ftarget, Ninitial, Nendo, and Fex are all milligrams per kilogram. The exogenous fertilizer requirement represents the amount of available soil nutrients that need to be supplemented from external sources through the application of bedding bio-organic fertilizer after deducting the soil's basal nutrient reserves and the release of nutrients from endogenous decomposition. When Ftarget is less than or equal to the sum of Ninitial and Nendo, the exogenous fertilizer requirement is zero or negative, indicating that the soil's natural nutrient supply capacity is sufficient to meet the nutrient needs of the next crop of pasture. In this case, no bedding bio-organic fertilizer is needed, and the minimum application rate is zero.

[0059] The seasonal mineralization rate of bedding bio-organic fertilizer is used as a mineralization release characteristic. The seasonal mineralization rate of bedding bio-organic fertilizer refers to the percentage of effective nutrients released through microbial mineralization during the next pasture growing season after the fertilizer is applied to the soil, relative to the total content of that nutrient in the bedding bio-organic fertilizer. Bedding bio-organic fertilizer originates from a cyclical process of straw bedding, fermentation bed aquaculture, and bedding composting. The seasonal mineralization rate of bedding bio-organic fertilizer is pre-determined using either an indoor aerobic culture method or a field nylon mesh bag method. Taking the indoor aerobic culture method as an example: a sample of bedding bio-organic fertilizer is mixed with soil from the target plot at, for example, a 1:100 weight ratio. The mixture is then aerobically cultured for one pasture growing season under conditions such as, for example, 25 degrees Celsius and 60% field water holding capacity. After the culture is completed, the net increase in effective nutrients in the soil of the mixed system is measured. The net increase in effective nutrients is divided by the total content of that nutrient in the bedding bio-organic fertilizer sample to obtain the seasonal mineralization rate of the bedding bio-organic fertilizer. The seasonal mineralization rate of bedding bio-organic fertilizer varies depending on the source and degree of composting. The seasonal mineralization rate of bedding bio-organic fertilizer is usually between 40% and 70%.

[0060] The minimum application rate is obtained by dividing the exogenous fertilizer requirement by the seasonal mineralization rate of the bedding bio-organic fertilizer. The minimum application rate is calculated as: Mmin = Fex / Rmin; where Mmin represents the minimum application rate, Fex represents the exogenous fertilizer requirement, and Rmin represents the seasonal mineralization rate of the bedding bio-organic fertilizer. Mmin and Fex are both in milligrams per kilogram. The minimum application rate represents the minimum amount of bedding bio-organic fertilizer required to ensure that the effective nutrient release from the bedding bio-organic fertilizer exactly compensates for the exogenous fertilizer requirement, assuming that the bedding bio-organic fertilizer can only release a portion of its nutrients in the current season. For example, when the exogenous fertilizer requirement is converted to, for example, 10 kg of alkaline nitrogen per acre, and the seasonal mineralization rate of the bedding bio-organic fertilizer is, for example, 50%, the minimum application rate is converted to, for example, 20 kg of alkaline nitrogen per acre of bedding bio-organic fertilizer. The method for converting the minimum application rate from nutrient content to the actual amount of bedding bio-organic fertilizer is as follows: divide the minimum application rate by the percentage content of that nutrient in the bedding bio-organic fertilizer. For example, if the percentage content of alkaline available nitrogen in the bedding bio-organic fertilizer is measured to be, for example, 1.5%, then the actual amount of bedding bio-organic fertilizer corresponding to 20 kg of alkaline available nitrogen per acre is approximately 20 / 0.015 = 1333 kg per acre.

[0061] Apply the minimum application rate. Use the calculated minimum application rate as the amount of bedding bio-organic fertilizer to be applied in the current season. Apply it evenly to the soil surface of the target plot in one go before the next crop of pasture is sown or during the seedling stage. After application, use shallow rotary tillage to mix the bedding bio-organic fertilizer evenly with the topsoil to ensure even distribution and sufficient contact with soil microorganisms, which is beneficial for the continuous supply of mineralized nutrients. The minimum application rate is used instead of the conventional recommended rate because the soil's basal nutrient holdings and endogenous decomposition release have already been fully deducted as basic nutrient sources; only the insufficient portion needs to be supplemented. Using the minimum application rate ensures that the predicted fertility is not lower than the target nutrient requirement, while avoiding over-application of bedding bio-organic fertilizer due to repeated calculations of endogenous decomposition release. This reduces the pressure of returning livestock waste to the field and minimizes the accumulation of excess nutrients in the soil and their release into the environment.

[0062] Example 2: Figure 2 A schematic diagram of a forage planting organic fertilizer application system based on fertility prediction is provided. The system includes: The sequence acquisition module is used to acquire the diurnal variation sequence of soil cellulase activity after harvesting of the target plot; The jump identification module is used to identify whether there are active jump points in the daily change sequence that exceed the range of daily average background fluctuations; The monitoring and acquisition module is used to acquire, if present, the soil effective nutrient time series monitoring value of the planting area as the first time series monitoring value, starting from the time corresponding to the active jump point, and acquire the soil effective nutrient time series monitoring value of the control area without root stubble residue as the second time series monitoring value. The correction processing module is used to calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, the difference curve is obtained by subtracting the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value. Based on the difference curve, the corrected initial rising slope and the time point corresponding to the curvature maximum value are obtained. The source apportionment module is used to input the corrected initial upward slope and the time point corresponding to the curvature maximum into the source apportionment relation to obtain the endogenous humic release and the soil background nutrient holdings; The application determination module is used to determine the minimum application amount that will ensure the predicted fertility of the next crop is not lower than the set standard, based on the soil's background nutrient holdings as the initial value, the amount of endogenous humification release, and the mineralization release characteristics of the bio-organic fertilizer to be applied.

[0063] All calculations involved in the embodiments are performed using dimensionless numerical values, and the preset parameters and thresholds in the calculations can be set by those skilled in the art according to actual conditions.

[0064] This technical solution can be flexibly deployed, for example, as embedded software running on device hardware, or installed on personal computers or other smart terminals with user interfaces, thus adapting to various hardware environments and usage requirements.

[0065] The above solutions can be implemented in software, hardware, firmware, or a combination thereof. When implemented in software, they are presented, in whole or in part, as a computer program product, containing one or more computer instructions or programs. When these instructions or programs are loaded and executed on a computer, results are produced corresponding to the processes or functions of the embodiments of this application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one medium to another via wired or wireless means, such as from a website, server, or data center via wired means such as fiber optic cables, twisted-pair cables, or coaxial cables, or wireless means such as infrared or microwaves to another site. A computer-readable storage medium refers to any usable medium that a computer can access or a data storage device such as a server or data center that contains one or more usable media, including magnetic media such as floppy disks, hard disks, and magnetic tapes, optical media such as DVDs, and semiconductor media such as solid-state drives.

[0066] The specific working process of the system, device and module can be found in the method embodiment, and will not be repeated here.

[0067] The disclosed systems, devices, and methods can be implemented in other ways. The device embodiments are for illustrative purposes only, and the module division is only a logical division. In practice, different divisions can be implemented, such as merging or integrating multiple modules or components, or omitting some features. Coupling, direct coupling, or communication connections between the components can be achieved through interfaces, while indirect coupling or communication connections can take electrical, mechanical, or other forms.

[0068] The modules described as separate components may or may not be physically separated. The components shown as modules can be physical hardware or software, and can be deployed centrally or distributed across multiple network nodes. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0069] The functional modules in each embodiment can be integrated into one processing module, or they can exist independently, or two or more modules can be integrated into one.

[0070] If the functionality is implemented as a software module and used as an independent product, it can be stored in a computer-readable storage medium. Under this understanding, the substantial contribution of the technical solution of this application can be embodied in the form of a software product. This computer software product is stored in a storage medium and contains instructions to cause a computer device, such as a personal computer, server, or network device, to execute all or part of the steps of the methods in the embodiments of this application. The storage medium includes any medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory, random access memory, magnetic disk, or optical disk.

[0071] The above are merely specific embodiments of this application, and the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should fall within the scope of protection of this application.

Claims

1. A method for applying organic fertilizer to forage crops based on fertility prediction, characterized in that, include: S1: Obtain the diurnal variation sequence of soil cellulase activity after harvesting in the target plot; S2: Identify whether there are activity jump points in the daily activity variation sequence that exceed the range of daily average background fluctuations; S3: If it exists, then starting from the time corresponding to the active jump point, obtain the soil effective nutrient time-series monitoring value of the planting area as the first time-series monitoring value, and obtain the soil effective nutrient time-series monitoring value of the control area without root stubble residue as the second time-series monitoring value. S4: Calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, subtract the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value to obtain the difference curve. Based on the difference curve, obtain the corrected initial rising slope and the time point corresponding to the maximum curvature. S5: Input the corrected initial upward slope and the time point corresponding to the curvature maximum value into the source analysis relation to obtain the endogenous decomposition release and the soil background nutrient holding; S6: Using the soil's baseline nutrient holdings as the initial value, and based on the amount of endogenous decomposition and release, combined with the mineralization and release characteristics of the bio-organic fertilizer to be applied, determine the minimum application amount to ensure that the predicted fertility of the next crop is not lower than the set standard, and then apply it.

2. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, Obtain the diurnal variation sequence of soil cellulase activity after harvesting in the target plot, including: After harvesting in the target plot, soil samples of the stubble residue layer were collected at preset time steps. The activity of soil cellulase in soil samples was measured to obtain the soil cellulase activity value at a single sampling time. Soil cellulase activity values ​​from multiple consecutive single sampling times are arranged in chronological order to form a diurnal variation sequence of soil cellulase activity.

3. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, Identify whether there are activity jump points in the daily activity variation sequence that exceed the daily average background fluctuation range, including: Soil cellulase activity values ​​were selected from the diurnal variation sequence of soil cellulase activity at the first few sampling times after harvest, and the average value was calculated as the daily background value. The standard deviation of soil cellulase activity values ​​at the first few sampling times after harvest was calculated as the background fluctuation range. The sum of the daily average background value and the background fluctuation amplitude is used as the threshold for determining the jump. The soil cellulase activity values ​​at subsequent sampling times in the diurnal variation sequence of soil cellulase activity are compared one by one with the jump threshold. When the soil cellulase activity value at a certain sampling time exceeds the jump threshold for the first time, the time point corresponding to that sampling time is identified as the activity jump point.

4. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, If present, starting from the time corresponding to the activity jump point, obtain the soil available nutrient time-series monitoring values ​​of the planting area as the first time-series monitoring value, and obtain the soil available nutrient time-series monitoring values ​​of the control area without root stubble residue as the second time-series monitoring value, including: Designate planting areas within the target plots, and preserve the root stubble remaining after harvesting within the planting areas; Within the target plot, a control area with no stubble residue was delineated, and all stubble residue remaining after harvesting in the control area with no stubble residue was removed. Starting from the time corresponding to the activity jump point, soil samples from the planting area and soil samples from the control area without root stubble residue are collected synchronously according to the preset time step. The available nutrient content of soil samples from the planting area was measured, and the available nutrient content of soil at each sampling time was arranged in chronological order to obtain the first time-series monitoring value. The available nutrient content of soil samples from the control area without stubble residue was measured, and the available nutrient content of soil at each sampling time was arranged in chronological order to obtain the second time-series monitoring value.

5. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, Calculate the ratio of the initial upward slope of the first time-series monitoring value to the background mineralization slope of the second time-series monitoring value. If the ratio is within a preset range, subtract the nutrient value at the corresponding time of the second time-series monitoring value from the first time-series monitoring value to obtain a difference curve. Based on the difference curve, obtain the corrected initial upward slope and the time point corresponding to the curvature maximum, including: Linear fitting is performed on the nutrient values ​​at the first few sampling times from the time corresponding to the activity jump point in the first time series monitoring value, and the slope of the fitted line is used as the initial rising slope of the first time series monitoring value. Linear fitting was performed on the nutrient values ​​at the first few sampling times from the time corresponding to the activity jump point in the second time series monitoring value, and the slope of the fitted line was used as the background mineralization slope of the second time series monitoring value. Calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value; Determine whether the ratio of the initial upward slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value is within a preset range; If it is within the preset range, the nutrient value at each sampling time in the first time series monitoring value is subtracted from the nutrient value at the corresponding sampling time in the second time series monitoring value to obtain the difference curve; Calculate the first-order difference sequence between each adjacent sampling time on the difference curve, identify the starting point of the segment in the first-order difference sequence where multiple consecutive negative values ​​first appear, and take the starting point of this segment as the critical sampling time for corrosion decay. The difference curve segment between the time corresponding to the activity jump point and the critical sampling time of corrosion decay is taken as the corrosion-dominant segment. The nutrient values ​​at the initial sampling times in the decomposition-dominant segment are linearly fitted, and the slope of the fitted line is used as the corrected initial upward slope. The curvature of the corrosion-dominant segment is calculated to obtain the sampling time when the curvature of the corrosion-dominant segment reaches its maximum value. This sampling time is taken as the time point corresponding to the maximum curvature value.

6. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 5, characterized in that, The process involves calculating the first-order difference sequence between adjacent sampling times on the difference curve, identifying the starting point of the segment in the first-order difference sequence where multiple consecutive negative values ​​first appear, and using this starting point as the critical sampling time for decomposition decay. This includes: arranging the nutrient values ​​at each sampling time on the difference curve in chronological order; calculating the difference between the nutrient value at the next sampling time and the nutrient value at the previous sampling time, arranging the differences of all adjacent sampling times in chronological order to obtain the first-order difference sequence; traversing the first-order difference sequence sequentially from its starting position, and when a certain position is reached, checking whether several consecutive difference values ​​from that position are all negative; if several consecutive difference values ​​from that position are all negative, then the sampling time corresponding to that position is identified as the starting point of the segment where multiple consecutive negative values ​​first appear, and is used as the critical sampling time for decomposition decay.

7. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, By inputting the corrected initial upward slope and the time points corresponding to the curvature maxima into the source apportionment relation, the endogenous decomposition release and soil background nutrient holdings are obtained, including: Obtain the pre-calibrated, corrected initial upward slope and the intrinsic decay release as the source resolution relationship; Input the corrected initial rising slope into the source analysis relation to obtain the intrinsic corruption exponent; Obtain the available soil nutrient content at the time point corresponding to the maximum curvature value in the control area without root stubble residue; The soil available nutrient content at the time point corresponding to the maximum curvature in the control area without root stubble residue was used as the soil background nutrient holdings.

8. The method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 7, characterized in that, To obtain the correspondence between the pre-calibrated corrected initial upward slope and the amount of endogenous humic release as the source apportionment relationship, the following steps are taken: selecting a calibration plot with the same soil type as the target plot and setting multiple calibration treatments with different root residue amounts; performing steps S1 to S4 on each calibration treatment to obtain the corrected initial upward slope corresponding to each calibration treatment; measuring the total net increase in available soil nutrients for each calibration treatment during the calibration period as the amount of endogenous humic release corresponding to each calibration treatment; and correlating the corrected initial upward slope corresponding to each calibration treatment with the amount of endogenous humic release corresponding to each calibration treatment to establish the correspondence between the corrected initial upward slope and the amount of endogenous humic release as the source apportionment relationship.

9. A method for applying organic fertilizer to forage grass cultivation based on fertility prediction according to claim 1, characterized in that, Using the soil's baseline nutrient levels as an initial value, and based on the amount of endogenous decomposition and release, combined with the mineralization and release characteristics of the bio-organic fertilizer to be applied to the bedding material, determine and apply the minimum application rate to ensure that the predicted fertility of the next crop is not lower than the set standard, including: The target fertilizer requirement for obtaining the next crop of forage is used as the standard for setting; The soil's baseline nutrient levels are used as the initial fertilizer supply; the amount released from endogenous humification is used as the endogenous fertilizer supply. The exogenous fertilizer requirement is obtained by subtracting the initial fertilizer supply from the target fertilizer requirement and then subtracting the endogenous fertilizer supply. The seasonal mineralization rate of the bedding bio-organic fertilizer was used as a mineralization release characteristic. The minimum application rate is obtained by dividing the required amount of exogenous fertilizer by the current season's mineralization rate of the bio-organic fertilizer in the bedding material. Apply at the minimum recommended dosage.

10. A fertility prediction-based organic fertilizer application system for forage planting, used to implement the fertility prediction-based organic fertilizer application method for forage planting as described in any one of claims 1-9, characterized in that, include: The sequence acquisition module is used to acquire the diurnal variation sequence of soil cellulase activity after harvesting of the target plot; The jump identification module is used to identify whether there are active jump points in the daily change sequence that exceed the range of daily average background fluctuations; The monitoring and acquisition module is used to acquire, if present, the soil effective nutrient time series monitoring value of the planting area as the first time series monitoring value, starting from the time corresponding to the active jump point, and acquire the soil effective nutrient time series monitoring value of the control area without root stubble residue as the second time series monitoring value. The correction processing module is used to calculate the ratio of the initial rising slope of the first time series monitoring value to the background mineralization slope of the second time series monitoring value. If the ratio is within a preset range, the difference curve is obtained by subtracting the nutrient value at the corresponding time of the second time series monitoring value from the first time series monitoring value. Based on the difference curve, the corrected initial rising slope and the time point corresponding to the curvature maximum value are obtained. The source apportionment module is used to input the corrected initial upward slope and the time point corresponding to the curvature maximum into the source apportionment relation to obtain the endogenous humic release and the soil background nutrient holdings; The application determination module is used to determine the minimum application amount that will ensure the predicted fertility of the next crop is not lower than the set standard, based on the soil's background nutrient holdings as the initial value, the amount of endogenous humification release, and the mineralization release characteristics of the bio-organic fertilizer to be applied.

Images